Skip to main content
SpringerLink
Log in

Fast load balancing with the most to least loaded policy in dynamic networks

  • Published:
The Journal of Supercomputing Aims and scope Submit manuscript

Abstract

Load balancing a distributed/parallel system consists in allocating work (load) to its processors so that they have to process approximately the same amount of work or amounts in relation with their computation power. In this paper, we present a new distributed algorithm that implements the Most to Least Loaded (M2LL) policy. This policy aims at indicating pairs of processors, that will exchange loads, taking into account actually broken edges as well as the current load distribution in the system. The M2LL policy fixes the pairs of neighboring processors by selecting in priority the most loaded and the least loaded processor of each neighborhood. Our first and main result is that the M2LL distributed implementation terminates after at most (n/2)⋅d t iterations where n and d t are respectively the number of nodes and the degree of the system at time t. We then present a performance comparison between Generalized Adaptive Exchange (GAE) that uses M2LL and Relaxed First Order Scheme (RFOS), two load balancing algorithms for dynamic networks in which only link failures are considered. The comparison is carried out on a dedicated test bed that we have designed and implemented to this end. Our second important result is that although generating more communications, the GAE algorithm with the M2LL policy is faster than RFOS in balancing the system load. In addition, GAE M2LL is able to achieve a more stable balanced state than RFOS and scales well.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Similar content being viewed by others

References

  1. Boillat JE (1990) Load balancing and Poisson equation in a graph. Concurr: Pract Exp 2(4):289–313

    Article  Google Scholar 

  2. Bahi JM, Couturier R, Vuillemin P (2006) Solving nonlinear wave equations in the grid computing environment: an experimental study. J Comput Acoust 14(1):113–130

    Article  Google Scholar 

  3. Willebeek-LeMair MH, Reeves AP (1990) Local vs global strategies for dynamic load balancing. In: Proc of the int conference on parallel processing, vol 1, pp 569–570

  4. Kumar V, Ananth GY, Rao VN (1991) Scalable load balancing techniques for parallel computers. Technical Report, pp 91–55. Dept of Computer Science, University of Minnesota, USA

  5. Willebeek-LeMair MH, Reeves AP (1993) Strategies for dynamic load balancing on highly parallel computers. IEEE Trans Parallel Distrib Syst 4(9):979–993

    Article  Google Scholar 

  6. Cortes A, Ripoll A, Senar MA, Luque E (1999) Performance comparison of dynamic load-balancing strategies for distributed systems. In: IEEE Proc of the 32th Hawai Int Conference on System Sciences, vol 8, pp 8041–8051

  7. Cybenko G (1989) Dynamic load balancing for distributed memory multiprocessors. J Parallel Distrib Comput 7(2):279–301

    Article  Google Scholar 

  8. Xu CZ, Lau FCM (1994) Optimal parameters for load balancing with the diffusion method in mesh networks. Parallel Process Lett 4(2):139–147

    Article  Google Scholar 

  9. Diekmann R, Frommer A, Monien B (1999) Efficient schemes for nearest neighbor load balancing. Parallel Comput 25(7):789–812

    Article  MathSciNet  Google Scholar 

  10. Elsasser R, Monien B, Preis R (2002) Diffusion schemes for load balancing on heterogeneous networks. Theory Comput Syst 35:305–320

    Article  MathSciNet  Google Scholar 

  11. Rotaru T, Nageli HH (2004) Dynamic load balancing by diffusion in heterogeneous systems. J Parallel Distrib Comput 64:481–497

    Article  MATH  Google Scholar 

  12. Hosseini SH, Litow B, Malkawi M, McPherson J, Vairavan K (1990) Analysis of a graph coloring based distributed load balancing algorithm. J Parallel Distrib Comput 10(2):160–166

    Article  Google Scholar 

  13. Xu CZ, Lau FCM (1992) Analysis of the generalized dimension exchange method for dynamic load balancing. J Parallel Distrib Comput 16(4):385–393

    Article  MATH  MathSciNet  Google Scholar 

  14. Aiello W, Awerbuch B, Zkfaggs B, Rao S (1993) Approximate load balancing on dynamic and asynchronous networks. In: Proc of the 25th annual ACM symposium on theory of computing, May, pp 632–641

  15. Bahi JM, Gaber J (2001) Load balancing on networks with dynamically changing topology. In: Proc of the 7th int Euro–par conf. on parallel processing, August 28–31, Manchester, Lecture Notes on Computer Science, pp 175–182

  16. Elsasser R, Monien B, Schamberger S (2004) Load balancing in dynamic networks. In: I–SPAN, 2004

  17. Bahi JM, Couturier R, Vernier F (2003) Accelerated diffusion algorithms on general dynamic networks. Proc of 5th int conference, PPAM Czestochowa, Poland. LNCS, vol 3019. Springer, Heidelberg, pp 77–82

    Google Scholar 

  18. Bahi JM, Couturier R, Vernier F (2003) Broken edges and dimension exchange algorithm on hypercube topology. In: Proc of the 11th Euromicro conference on parallel, distributed and network-based processing (Euro-PDP’03)

  19. Bahi JM, Couturier R, Vernier F (2005) Synchronous distributed load balancing on dynamic networks. J Parallel Distrib Comput 65(11):1397–1405

    Article  MATH  Google Scholar 

  20. Bahi JM, Couturier R, Sider A (2006) Design and analysis of the M2LL policy distributed algorithm for load balancing in dynamic networks. Proc of the 2006 int symp on parallel and distributed processing and applications (ISPA’06). LNCS, vol 4331. Springer, Heidelberg, pp 195–204

    Google Scholar 

  21. Xu CZ, Monien B, Lüling R, Lau FCM (1995) Nearest neighbor algorithms for load balancing in parallel computers. Concurr: Practice Exp 7:707–736

    Article  Google Scholar 

  22. Fiorini S, Wilson RJ (1978) Edge-coloring of graphs. In: Beineke LW, Wilson RJ (eds) Selected topics in graph theory. Academic Press, New York

    Google Scholar 

  23. Vernier F (2004) Algorithmique itérative pour l’équilibrage de charge dans les réseaux dynamiques. PhD thesis, Université de Franche-Comté, France

Download references

Author information

Authors and Affiliations

  1. Département d’Informatique, Université Abderrahmane Mira de Béjaia, Route de Targa Ouzemmour, Béjaia, 06000, Algérie

    Abderrahmane Sider

  2. Laboratoire d’Informatique de l’Université de Franche-Comté(LIFC), IUT de Belfort-Montbéliard, University of Franche-Comte, BP 527, 90016, Belfort Cedex, France

    Raphaël Couturier

Authors
  1. Abderrahmane Sider
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Raphaël Couturier
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Abderrahmane Sider.

Rights and permissions

Reprints and permissions

About this article

Cite this article

Sider, A., Couturier, R. Fast load balancing with the most to least loaded policy in dynamic networks. J Supercomput 49, 291–317 (2009). https://doi.org/10.1007/s11227-008-0238-5

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11227-008-0238-5

Keywords

Search

Navigation